Laboratory adaptation of the SARS-CoV-2 Delta variant in Vero E6 cells enhances in vitro growth but attenuates human TMPRSS2-dependent fitness, pathogenicity, and transmission

This is the authors' abstract. We don't add key points for this paper.

Laboratory propagation of severe acute respiratory syndrome coronavirus 2 is essential for virological research, but adaptation in commonly used non-human cells with low transmembrane serine protease 2 expression may alter phenotypes relevant to biosafety and risk assessment. This study aimed to determine how serial passage in African green monkey-derived Vero E6 cells affects the growth, entry pathways, pathogenicity, and transmission of a clinical Delta strain. The parental virus was serially passaged to generate two adapted variants, which were characterized by whole-genome sequencing, replication and competition assays, spike cleavage and membrane fusion analyses, receptor binding studies, and animal experiments. The first variant acquired spike substitutions P217S, S247R, and H655Y, whereas the second acquired G261R and a six-amino-acid deletion immediately upstream of the furin cleavage site, plus non-spike mutations. Both variants replicated more efficiently than the parental virus in Vero E6, LLC-MK2, Huh-7, and HeLa cells expressing human angiotensin-converting enzyme 2 (ACE2), but not in Caco-2 cells. Their phenotypes reflected altered use of human transmembrane serine protease 2 rather than altered human ACE2 binding; they showed increased affinity for African green monkey ACE2 but unchanged affinity for the human receptor. Despite enhanced growth in vitro , both variants exhibited reduced pathogenicity and transmission in animals. These findings demonstrate that Vero E6 passage can rapidly reshape spike function and select viruses with improved cell-culture growth but impaired protease-dependent fitness in vivo , supporting rigorous genomic and phenotypic monitoring of laboratory-propagated virus stocks before use.